A Survey of Image Gradient Inversion Against Federated Learning

In order to preserve data privacy while fully utilizing data from different owners, federated learning is believed to be a promising approach in recent years. However, aiming at federated learning in the image domain, gradient inversion techniques can reconstruct the input images on pixel-level only by leaked gradients, without accessing the raw data, which makes federated learning vulnerable to the attacks. In this paper, we review the latest advances of image gradient inversion techniques and evaluate the impact of them to federated learning from the attack perspective. We use eight models and four datasets to evaluate the current gradient inversion techniques, comparing the attack performance as well as the time consumption. Furthermore, we shed light on some important and interesting directions of gradient inversion against federated learning.<br>

A Survey of Image Gradient Inversion Against Federated Learning

In order to preserve data privacy while fully utilizing data from different owners, federated learning is believed to be a promising approach in recent years. However, aiming at federated learning in the image domain, gradient inversion techniques can reconstruct the input images on pixel-level only by leaked gradients, without accessing the raw data, which makes federated learning vulnerable to the attacks. In this paper, we review the latest advances of image gradient inversion techniques and evaluate the impact of them to federated learning from the attack perspective. We use eight models and four datasets to evaluate the current gradient inversion techniques, comparing the attack performance as well as the time consumption. Furthermore, we shed light on some important and interesting directions of gradient inversion against federated learning.<br>

Shallow Water Bathymetry Inversion Using Hybrid Multi-band Composites Resulted from Integration of Drone-based RGB and Multispectral Imagery

Shallow bathymetry inversion algorithms have long been applied in various types of remote sensing imagery with relative success. However, this approach requires that imagery with increased radiometric resolution in the visible spectrum is available. The recent developments in drones and camera sensors allow for testing current inversion techniques on new types of datasets. This study explores the bathymetric mapping capabilities of fused RGB and multispectral imagery, as an alternative to costly hyperspectral sensors. Combining drone-based RGB and multispectral imagery into a single cube dataset, provides the necessary radiometric detail for shallow bathymetry inversion applications. This technique is based on commercial and open-source software and does not require input of reference depth measurements in contrast to other approaches. The robustness of this method was tested on three different coastal sites with contrasting seafloor types. The use of suitable end-member spectra which are representative of the seafloor types of the study area and the sun zenith angle are important parameters in model tuning. The results of this study show good correlation (R2&gt;0.7) and less than half a meter error when they are compared with sonar depth data. Consequently, integration of various drone-based imagery may be applied for producing centimetre resolution bathymetry maps at low cost for small-scale shallow areas.

Shallow water bathymetry inversion using hybrid multi-band composites resulted from integration of drone-based RGB and multispectral imagery

Shallow bathymetry inversion algorithms have long been applied in various types of remote sensing imagery with relative success. However, this approach requires that imagery with increased radiometric resolution in the visible spectrum is available. The recent developments in drones and camera sensors allow for testing current inversion techniques on new types of datasets. This study explores the bathymetric mapping capabilities of fused RGB and multispectral imagery, as an alternative to costly hyperspectral sensors. Combining drone-based RGB and multispectral imagery into a single cube dataset, provides the necessary radiometric detail for shallow bathymetry inversion applications. This technique is based on commercial and open-source software and does not require input of reference depth measurements in contrast to other approaches. The robustness of this method was tested on three different coastal sites with contrasting seafloor types. The use of suitable end-member spectra which are representative of the seafloor types of the study area and the sun zenith angle are important parameters in model tuning. The results of this study show good correlation (R2&gt;0.7) and less than half a meter error when they are compared with sonar depth data. Consequently, integration of various drone-based imagery may be applied for producing centimetre resolution bathymetry maps at low cost for small-scale shallow areas.

A Comprehensive Comparison of Period Extraction Algorithms for Asteroids with Long Term Observation

The light curve period of an asteroid plays an important role in determining the rotation period, the collision evolution and the YORP effect. There are many period extraction algorithms used to find the light curve period of asteroids with long term observation, which are mainly based on the frequency, time and time–frequency domains. This paper presents a comprehensive and unparalleled comparison of the popular algorithms based on the DAMIT (Database of Asteroid Models from Inversion Techniques) data set to show the statistical results. Considering the quoted period, absolute magnitude, diameter, albedo, time span and number of observations, we analyze the accuracy of five popular methods using the light curve data of 2902 asteroids. We find that although the performance of all the algorithms varies little, Phase Dispersion Minimization (PDM) performs better, followed by Lomb-Scargle (LS), while Conditional Entropy (CE) is not better than the others under certain conditions. We also analyze the cases which are more suitable for searching by frequencies or by periods.

Identification of Crustal Thickness in Central Part of Sumatra Using Teleseismic Receiver Function Method

The Central part of Sumatra is a region that has a high potential for earthquakes. This research intended to determine the crustal thickness of the earth, P and S wave velocity profiles, and vp/vs value in the Central part of Sumatra using stacking H-k and inversion techniques based on the analysis of receiver function. This study utilized teleseismic earthquake data with a distance of 30° to 90° from the station and magnitude more than 6 (M>6). The stations used in this study were 3 BMKG broadband stations located in 3 zones, the fore arc ridge zone (SISI), the volcanic zone (PLSI) and the back arc zone (TPRI). The crustal thickness varies in the fore arc ridge zone (SISI) estimated 17.8 km, volcanic zone (PLSI) reaches 29.7 km and the back arc zone (TPRI) reaches 34 km. The crustal thickness is quite thick under the PLSI and thicker beneath TPRI station. These possibly due to the effect of topography and isostatic compensation in the station. However, whether there is a correlation between crustal thickness and topography needs further research using more stations. The highest vp/vs value was found in the volcanic zone of 1.9, that might be associated with the presence partial melting beneath the station. Meanwhile, the vp/vs value in the back arc zone is 1.72, indicating a relatively more homogeneous structure.

Empirical Modelling of Public Lighting Emission Functions

Study of light at night has increased in recent decades due to the recognition of its impact on the environment, potential health concerns, as well as both the financial and carbon cost of energy waste. The advent of more extensive and improved ground-based measurements together with quantifiable satellite data has revolutionised the field, and provided data to test improved theoretical models. However, “closing the loop” and finding a detailed connection between these measurements requires knowledge of the “city emission function”, the angular distribution of upwelling radiation with zenith distance. Simplified analytical functions have been superseded by more complex models involving statistical approximation of emission sources and obstructions and inversion techniques now permit the estimation of emission functions from the observed sky brightness measurements. In this paper, we present an efficient GIS-based method to model public lighting using real-world photometric data and high-resolution digital elevation maps of obstructions such as buildings and trees at a 1 m scale. We discuss the results of this work for a sample of Irish towns as well as a city area. We also compare our results to previous emission functions as well as to observed asymmetries in emission detected by satellites such as SUOMI VIIRS.